David B. Benson wrote:With your paper tubes maybe no library paste is required, which would be better.
During preparation of the drawing, it became obvious that the key to this whole affair is in the floor connections. Their composition, placement, and means of adherence dictate the dynamics and outcome of the experiment. Indeed, the importance was spelled out in your opening post:
...let go, and observe the crush-down (unless the connectors are too big).
or too strong or well-attached or inadequately placed. The rest of the choices for structural components are all a matter of expediency. To be honest, in any physical model I do build, it will end up being made with the Legos I have in buckets downstairs or another (cheaper) material. I've already pumped enough money into Lego Inc's wallet!
Anyway, the matter of connectors is quite serious. I understand a first approximation ignoring the walls, but this demands extra scrutiny on the connections, as this will be a connection-only failure mode. As such, it has an equivalence relation to pancaking and, at even higher levels of abstraction, houses of cards and domino chains. Unlike pure pancaking, the proposed model adds the significant mass and rigidity of the walls to drive the pancaking, the lower walls are merely to contain the rigid avalanche in a chute, as I understand it. I believe this is at least part of the point Heiwa was trying to get at when inquiring about the use of toilet paper.
Oops. Not quite right. All playing cards are to be exactly the same size.
Ideally, yes, but in a practical sense I don't think it matters too much, if the motivation is uniform slab mass and geometry. I think the upper walls dimensioned inside the lower walls renders this concern moot. Please correct me if I'm missing the point. Again, it seems the properties of the connection dominate the concerns, and varying the connector length as opposed to slab dimension could introduce more undesirable distinction in block behavior.
I chose tubes because they're easy and they can give variable failure energies per unit, though that's certainly not the only way it could be done, probably not the best way. In the renderings I've shown, the tube is depicted laying across a wall brick inside an opening with available vertical clearance. No doubt, in this configuration, some sort of adhesive is required to keep it in place. Given the torques involved for even cardstock (really can't use playing cards unless it's a jumbo novelty deck, the scale needs to be greater), a strong attachment is required to satisy any reasonable DCR in a building analogy. My only experience with library paste is throwing baggies of it against some unlikely targets, and I remember wondering what sort of a problem it would be to get it off after it dried.
Whatever the means of tube attachment or other connection scheme, it should primarily satisfy the constraint of easy re-use, as I suspect many trials will be necessary. Library paste, while soluble, may not be the easiest in this regard. Drying times are a factor, too. All of this suggests the search for a non-degradable, adjustable and purely mechanical means of connector failure is probably worth the effort.
It's not my intention to introduce inordinate complexity with simulations, upfront analysis and multiple trials in what is otherwise a simple table-top demonstration by design. The simplicity is the hook. However, just the fact that the connections can be 'tuned' to fail or survive warrants further examination in order to bring out the most in the model. The really interesting thing for me is to find, in the physical model, the threshold of arrest and then admit that into the theoretical framework and see the clearer resulting picture. In my very simplified
crush-up simulacra, I've observed the extreme sensitivity about the collapse/arrest bifurcation. I suspect the only way a physical model will be able to resolve the transition region is through careful choice of participating dynamic constituents.
